The impact on climate change of the release of large amounts of CO₂ has attracted widespread attention [1], [2], and the challenges posed by global climate change have prompted a search for innovative solutions to reduce greenhouse gas emissions [3]. China’s dual carbon goals of achieving a carbon peak and carbon neutrality are essential in order to the nation to promote high-quality economic and social development, further advance the construction of an ecological civilization, and proactively respond to global climate change [4], [5]. In this context, China has successively formulated and implemented a series of carbon capture, utilization, and storage (CCUS) technology roadmaps and strategic plans [6], [7]. CCUS is regarded as an important strategic development direction, and a number of key technology research projects in CCUS are continuously being carried out, accelerating the breakthrough of core technologies and promoting the construction of an industry model. CO₂-enhanced oil recovery (EOR) is the most practicable CCUS technology at present and is the major technical support for China’s dual carbon goals; it is a significant contributor to the construction of a CCUS industry model that integrates oil recovery and storage [8].

In addition, China attaches great importance to energy, and the Chinese government has produced a series of important instructions, putting forward requirements such as “Greatly enhance the exploration and development efforts to ensure China’s energy security” and “Put the energy rice bowl in our own hands” [9]. Against the background of the gradual depletion of traditional oil resources, shale oil with its abundant reserves has gradually become an important part of China’s energy security [10], [11], [12]. However, shale reservoirs are characterized by low permeability, rapid reduction of production, and low primary recovery rate [13]. Although CO₂ injection is an effective technology to enhance the recovery of reservoirs, shale reservoirs are dense and non-homogeneous, and it is difficult for the traditional CO₂-EOR method to satisfy the requirements of unconventional oil reservoir development in China’s terrestrial facies [14].

CO₂ fracturing is a new type of CCUS-EOR technology that uses CO₂ as a fracturing medium to stimulate reservoirs. This technology can effectively liberate a large amount of shale oil resources, while simultaneously achieving the safe geological storage of CO₂—an extremely important technical support for realizing a carbon peak and carbon neutralization [15]. Compared with conventional CO₂ injection technology, CO₂ is injected into the reservoir at high pressure and high capacity during the fracturing stage of CO₂ fracturing, which makes it easier for CO₂ to enter nanoscale pore channels.

CCUS technology has played an important role worldwide in the exploration of unconventional reservoirs [16], [17]. Liquid CO₂ fracturing was first used in the oil industry in the early 1960s [18]. Palmer and Sito [19] demonstrated better performance with CO₂ fracturing fluids than with other fracturing fluids based on production data from 66 horizontal wells. Song et al. [20] found that CO₂ fracturing was 81.9% more effective than conventional fracturing, allowing for higher stimulated reservoir volume (SRV) and higher production. Due to the high-temperature and high-pressure conditions in shale reservoirs, CO₂ fracturing fluids exist in a supercritical (SC) state [21], [17]. Wu et al. [22] reported that SC-CO₂ could permeate fracture tips, quickly resulting in a dynamic propagation process and generating plenty of tensile fractures. Giesting et al. [23] found that SC-CO₂ can easily penetrate into tiny pores and fractures without any blocking effect. Therefore, it is of strategic importance to adopt CO₂ fracturing technology in unconventional oil and gas production in order to reduce CO₂ emissions and improve energy efficiency.

Similarly, CO₂ fracturing in shale reservoirs combined with numerical simulation holds great research value. Eshkalak et al. [24] conducted an evaluation of data from a shale gas field. They performed numerical simulations using local grid refinement (LGR) and a dual-porosity/dual-permeability model. Wang et al. [25] proposed a two-phase filtration-rate calculation model SC-CO₂ fracturing in unconventional natural gas reservoirs. Their model accounts for the impact of CO₂–shale–CH₄ adsorption. Zhao et al. [26] developed a model for CO₂ fracturing in naturally fractured reservoirs. Their findings demonstrate that CO₂ fracturing can generate longer but narrower fractures than traditional methods due to its lower breakdown pressure and pressure drop within the fractures. More specifically, the pressure drop in a single fracture during CO₂ fracturing is 60.48% and 78.88% lower than those of slick water and gel fracturing, respectively, when injected for the same duration. Fianu et al. [27] studied several temperature-dependent gas adsorption models and applied them to various shale gas reservoirs via simulations. Spanakos and Rigby [28] found that the adsorption of CO₂ molecules and the desorption of methane molecules increased by 2.74% and 2.30%, respectively, in regions with high surface capacity compared with a model that did not take surface diffusion into account. Wang et al. [29] considered a variety of mechanisms, including adsorption and diffusion, to describe CO₂ transport and storage in multi-stage fractured horizontal wells in shale reservoirs. Most reservoir parameters have a negligible effect on storage capacity at low injection pressures but have a significant effect on storage capacity at high injection pressures. Overall, commercial simulator platforms offer convenient tools for conducting CO₂ fracturing simulations in shale reservoirs based on these approaches.

Although experiments and simulations have been conducted to study the role of CO₂ as a fracturing fluid in shale reservoirs, most of these studies lack the support of field practice. There is also a lack of research on the storage capacity of CO₂ fracturing. The aim of this paper is to explore in deeper detail the application and significance of CO₂ fracturing technology in unconventional shale reservoirs and to reveal the potential and challenges of CCUS technology through lessons learned from global practices and China’s specific situation. In this paper, the role of CO₂ fracturing for CO₂ storage in shale reservoirs is simulated using the commercial simulator CMG–GEM in combination with field data. Predictions for the future are made based on historical matching. The feasibility and performance of CO₂ storage, hydraulic fracturing, and other reservoir engineering techniques in shale reservoirs are also evaluated. Fig. 1 illustrates the role of CO₂ in fracturing for CO₂ storage and EOR. It also predicts the future CO₂-storage potential of the Gulong shale oil reservoir in Daqing Oilfield. The research and application of CCUS technology are expected to contribute positively to global energy sustainability and climate change issues, as well as provide critical support for unconventional oil and gas development in China.

The Gulong Depression is a negative tectonic unit in the Songliao Basin, developed based on a basement tectonic pattern. The Gulong shale reservoir well 1 (GYYP1 well) is a horizontal well for the development of shale oil in the history of Daqing Oilfield; it also has important significance for Daqing Oilfield in enabling the application of technology that differs from that of a conventional oil reservoir. From August 24 to September 9, 2019, CO₂ fracturing technology was used at the GYYP1 well at the Daqing Oilfield to achieve large-scale stimulation. Since oil production first began at the GYYP1 well, the well’s oil and gas production has maintained steady growth with the gradual enlargement of the nozzle. The design was based on 36 stages of the fracturing process; however, due to obstruction of the fracturing process, the 26th stage was dropped, and the actual fracturing process occurred in 35 stages. Fig. 2 [30] shows a schematic diagram of the microseismic monitoring data indicating the fracture distribution for the GYYP1 well.

The CMG-GEM simulator is used in this study to characterize the transport dynamics of fluids [31]. This simulator was specifically designed for modeling and analyzing the behavior of unconventional reservoirs, including shale reservoirs. The CMG-GEM simulator incorporates advanced algorithms and numerical techniques to simulate complex processes such as fluid flow, adsorption, and diffusion in shale formations. A dual-permeability model was developed for shale reservoirs as the broad-scale SRV. CO₂ is stored in the shale mainly in its SC and adsorbed phases. Due to the short period of storage time in shale reservoir development, the impact of chemical reactions is not considered in the model. The reservoirs are assumed to be isothermal, with gas adsorption fitting the Langmuir isotherm equation [32].

where $B_i$ is the parameter for the Langmuir isotherm relation; ${\omega }_i$ is the number of moles of adsorbed component per unit mass of rock; ${\omega }_{i,\text{max}}$ is the maximum number of moles of the adsorbed component $i$ per unit mass of rock; $p$ is the pressure; $y_{i\text{g}}$ is the molar fraction of the adsorbed component $i$ in the gas phase; $j$ is the number of all components to be calculated in the summation function $\sum$. The diffusion coefficient of CO₂ is 5 × 10^–7 cm²·s⁻¹ [33], and CO₂ gas diffusion obeys the Sigmund method [34]:

where $D_{\mathit{ij}}$ is the diffusion coefficient; ${\rho }_k$ is the density and the subscript k denotes the phase; ${\rho }_k^0$ and $D_{\mathit{ij}}^0$ are respectively the initial density and diffusion coefficient; and ${\rho }_{k\text{r}}$ is the reduced density.

This study selected the target reservoir of the GYYP1 well to study its production history. Based on the geological conditions, the depth of the reservoir is set to 2584.9 m. The initial pressure is 32 MPa and the temperature is 112 °C. There are 100 × 50 grids in a single layer. Each grid is 20 m in length and 16 m in width. The porosity of the grid is 0.10, the permeability is 0.03 × 10⁻¹⁵ m², and the ratio of vertical permeability (K_v) to horizontal permeability (K_h) in the reservoir is 10. As the model uses structured grids, the transverse hydraulic fractures are created with a plane perpendicular to the x-direction. Thus, by applying logarithmic LGR around the fracture, the objective of modeling the flow through the fracture is achieved. The half-length of the fractures is set based on field microseismic detection data, which are shown in Fig. 3 according to the plan view. The GYYP1 well is a CO₂ injection well in the reservoir at the early stage, with a length of 1900 m and a total of 35 stage fractures perpendicular to the horizontal well. During the field fracturing construction, about 75 t of CO₂ and 2000 m³ of slick water on average were injected into each stage of the fractures. The average pressure of the ground monitoring during the injection of CO₂ was 45 MPa, while the pressure for the subsequent sand fracturing reached 60 MPa and above. After fracturing was completed, the GYYP1 well was turned into a production well for continued development.

To simulate production in the field, all the CO₂ was injected into the reservoir over a 17-day period, totaling approximately 3475 t along with 8.27 × 10⁴ m³ of water; this was followed by depletion production. After six months of production, gas began to be produced. Fig. 4 shows the results of the CO₂ production history matching for an additional three years after gas production. The numerical simulation results were similar to the historical data. The correlation coefficient for the history matching was greater than 0.96. After three years, the storage efficiency of CO₂ was 85.03%, which is in line with real field data. The numerical model was taken as a basis for predictions. In addition, the Duong method was used for the rate-decline analysis, helping to better predict the production. The Duong model has good application in shale reservoirs [35]; in addition, it was well matched to the amount of CO₂ production, with a correlation coefficient greater than 0.98. According to the Duong model, 704.06 t of CO₂ were produced after ten years.

The numerical simulation was extended to ten years to study the case of CO₂ storage. Fig. 5 depicts the injection and production mass of CO₂ during this ten-year period. Initially, in the first three years, the rate of CO₂ production significantly declined due to the high CO₂ saturation around the wellbore, resulting in a CO₂ storage rate of 85.03%. However, subsequent production showed a stable rate of CO₂ production, progressively decreasing from 150 m³·d⁻¹ in 2023 to 40 m³·d⁻¹ in 2029. After ten years, the cumulative amount of CO₂ stored in the subsurface through the GYYP1 well exceeded 2700 t, leading to an efficient CO₂ storage rate of 80.15%. By utilizing the CO₂ fracturing technique, the well has a high carbon-storage effect. These findings indicate that the Gulong Depression possesses immense potential for long-term CO₂ storage.

Fig. 6 illustrates the variations in the number of moles and the percentages of CO₂ for different storage mechanisms over a ten-year period. Initially, during the simulation, the injected CO₂ predominantly existed in a SC state, as it did not have sufficient time to react. However, as the injection process progressed, CO₂ readily combined with the shale formation, leading to an increase in adsorption. During the fracturing process, it was necessary to maintain very high pressure in the reservoir, so it was more favorable for t.0he CO₂ to adsorb onto the shale. Eventually, by the end of the fracturing process, the adsorption amount reached 22.13%. Subsequently, the GYYP1 well was converted into a production well to reduce the bottomhole pressure for production purposes. As the adsorption of CO₂ was closely and positively linked to the pressure, the adsorption amount gradually decreased during the production stage. This is also favorable to the recovery of SC-CO₂, thus resulting in a fast decrease of SC-CO₂ in reservoir. After ten years, the percentage of adsorbed CO₂ was found to be 11.06%.

Fig. 7 depicts the distribution of the adsorbed CO₂ over a ten-year period. The figure has been locally enlarged to show the 33rd–35th stages of the fracture closest to the wellhead. The CO₂ adsorption around the horizontal well was quite small at the beginning, not exceeding 10 mol·m⁻³. The CO₂ adsorption was mainly concentrated at the fractures, where the CO₂ adsorption concentration was as high as 1082 mol·m⁻³. With the diffusion of CO₂, the CO₂ concentration at the horizontal well gradually decreased to 30 mol·m⁻³. Upon CO₂ injection, the majority of the CO₂ was concentrated around the fractures and exhibited high content. However, as the pressure decreased, the adsorbed CO₂ underwent a conversion to its free state, leading to a decrease in concentration within the fracturing zone. Due to the exceedingly low permeability of shale reservoirs, the transfer of CO₂ to distant areas was challenging. Consequently, by the end of the production stage, the adsorbed CO₂ predominantly filled the vicinity surrounding the horizontal well.

Diffusion plays a crucial role in CO₂ storage within shale reservoirs, as it is closely linked to the shale’s characteristics and serves as an important parameter for adsorption analysis. Fig. 8 illustrates the CO₂ storage mechanism over a ten-year period, comparing scenarios with and without diffusion. The curves in the figure demonstrate that, during the initial stages of reservoir development, the influence of diffusion is not apparent. With freshly completed fracturing and elevated reservoir pressure, both diffusion-based and non-diffusion-based adsorption result in significant quantities of CO₂ being adsorbed, which subsequently decreases with declining pressure.

However, the impact of diffusion is evident after one year of production. Diffusion enhances the reaction of CO₂ upon contact with the shale surface, facilitating the transfer and exchange of substances. In the presence of diffusion, the amount of adsorbed CO₂ is considerably higher than in its absence. Moreover, this difference progressively increases over time. After ten years, the total adsorption with diffusion amounts to 6.15 × 10⁶ mol, in comparison with 4.88 × 10⁶ mol without diffusion, representing a relative increase of 26.02%.

Diffusion can also facilitate fluid flow in shale reservoirs, which can impact the transportation of oil and gas. In the absence of diffusion, any unadsorbed CO₂ remains in a supercritical state, making it more prone to reservoir production and resulting in storage losses.

To combat global climate change and contribute to the achievement of China’s dual carbon goals, the application of CO₂ fracturing holds great promise, as it not only enhances the flow capacity of shale reservoirs but also provides a means of storing CO₂ effectively. The GYYP1 well in Daqing Oilfield is a pioneering example of exploration in the development of unconventional shale reservoirs. The development plan for this well, utilizing CO₂ fracturing technology, has immense value and serves as a guiding reference for future endeavors in the oilfield.

Daqing Oilfield has set ambitious targets in its new construction capacity, with over 1000 new wells scheduled for completion in shale oil reservoirs by 2030. Building upon the experience gained from the GYYP1 well, the application of CO₂ fracturing technology for CO₂ storage has the potential to achieve remarkable results. With an average injection volume of 2500 t per well and a storage rate of 80%, a cumulative amount of approximately two million tons of CO₂ can eventually be stored. This achievement will significantly contribute to the goals of achieving a carbon peak and carbon neutrality.

The future focus of research and application in this field will be on the role of CO₂ in the development of unconventional shale reservoirs. Advancements made in these key areas will profoundly impact the sustainability and environmental friendliness of unconventional oil and gas development. Taking into account geomechanics and fracture extension before hydraulic fracturing simulation may help to get more accurate results [36], [37]. Aside from CO₂ fracturing, there are two other parts of a well’s life stage in which carbon storage is effective. Fig. 9 shows the CO₂ utilization in a shale reservoir at different well life stages, indicating the amount of potential CO₂ storage based on the degree of greenness. Firstly, it is possible to cause oil and gas production loss without an energy supply. Therefore, research on the CO₂ huff-n-puff process in unconventional shale reservoirs will be necessary [38], [39], [40]. This technology enables the efficient exploration of shale reservoirs through the continuous injection of CO₂, which not only enhances oil recovery but also reduces greenhouse gas emissions. It would be an economical and sustainable measure. However, several challenges still remain to be addressed in current CO₂ huff-n-puff technology, such as the selection of CO₂, the optimization of injection conditions, and the release of gas in shale reservoirs. Secondly, once the shale reservoirs become depleted, with low oil production, there is still a large potential for CO₂ storage by means of CO₂ injection [41], [42]. This stage will allow for the storage of the largest amount of CO₂. Some additional work is required to fully investigate the potential for CO₂ storage in shale reservoirs, including the evolution of CO₂ underground and the safety and effectiveness of CO₂ storage. Future research efforts could also aim to deepen the current understanding of CO₂ adsorption and desorption processes in shale reservoirs. These research findings will provide more effective solutions for the sustainable development of unconventional shale reservoirs, ensuring the protection of the environment and national energy security, as well as the achievement of China’s dual carbon goals.

This study delves into the potential of CO₂ storage by leveraging CO₂ fracturing technology in shale reservoirs. A horizontal well in the Gulong shale reservoir was taken as the study object, with its CO₂ storage impact being evaluated through data originating from the GYYP1 well. In this study, a dual-permeability fracturing model was established using CMG–GEM, matching the CO₂ production history, and subsequent production was projected for a ten-year period.

The results displayed a high CO₂ storage efficiency over a decade, reaching 80.15%—a promising indication of considerable storage potential. The absorbed CO₂ peaked at 22.13% during the initial fracturing procedure; however, as pressure diminished due to constant development, this ratio dropped to 11.06% after the ten-year period. Notably, most of the absorbed CO₂ was found to be concentrated around the horizontal well, which was due to the low permeability of the shale reservoir.

This study also considered the importance of diffusion in CO₂ storage theory. Diffusion increases the contact area between CO₂ and the rock; during the studied decade, the adsorption in the presence of diffusion increased by a relative 26.02% compared with that in the absence of diffusion. Thus, both diffusion and adsorption mechanisms play a significant role in CO₂ fracturing techniques in shale reservoirs, suggesting impressive CO₂ storage capabilities.

An extrapolation of the results from the GYYP1 well suggests that around 1000 future wells in Gulong shale oil reservoirs could harness similar storage potential, amounting to a storage capacity of nearly two million tons by 2030. If research continues to enhance CO₂ fracturing and storage methodologies, this technology holds substantial promise for contributing to energy security and pursuing China’s dual carbon goals.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was funded by the General Program of the National Natural Science Foundation of China (52274058, 52174052, and 52474058), the Central Program of Basic Science of the National Natural Science Foundation of China (72088101), and the “Enlisting and Leading” Science and Technology Project of Heilongjiang Province (RIPED-2022-JS-1740 and RIPED-2022-JS-1853).